All cells in the hematopoietic system originate from hematopoietic stem cells (HSCs) which are multipotent progenitor cells responsible for the hematopoietic cell lineage and are mainly found in peripheral blood and bone marrow (BM) (1). The multipotent progenitor pool of HSCs is heterogeneous and can be divided into long-term repopulating HSCs (LT-HSCs), short-term HSCs (ST-HSCs), and multipotent progenitors (MPPs) populations (1). LT-HSCs have a long-lasting capacity of self-renewal and differentiation, enable lifelong production of all hematopoietic lineages, and the replenish the hematopoietic system after injury or transplantation (2). ST-HSCs and MPPs that are differentiated from LT-HSCs have rapid proliferation and wide differentiation potential, but ST-HSCs have just a transient self-renewal ability and do not possess the same long-term self-renewal capacity as LT-HSCs (2).

HSC transplantation is a widely used treatment for various hematological disorders including leukemia (3, 4). Successful HSC transplantation depends on the ability of the transplanted HSC to maintain the hematopoietic BM niche and produce sufficient hematopoietic cells within the patient’s BM. Therefore, to maintain long-term hematopoiesis after transplantation, the role of LT-HSCs maintaining a hematopoietic niche for a long time is crucial. Here, we describe the mechanism of how LT-HSC is maintained in BM stem cell niche and then its clinical applications.

KAI1, also known as CD82 (KAI1 hereafter) belongs to the tetraspanin superfamily (5, 6) and has attracted attention because it has been shown to suppress cancer metastasis (7). KAI1 was initially identified as a possible indicator of anti-metastasis for prostate cancer. The metastasis suppressor role of KAI1 was firstly discovered through a genetic screening method using the metastatic rat AT6.1 prostate cancer cell line. The re-expression of KAI1 in AT6.1 cells led to a significant reduction in metastasis when these cells were injected subcutaneously into nude mice.

According to previous reports, KAI1 is widely expressed in various tissues and cell types, and its expression and structure are evolutionally conserved (5). KAI1, which was previously known as C33, 4F9, IA4, and R2 (8, 9), has a regular tetraspanin structure with N- and C-termini within cell membrane, as well as two extracellular loops (EC1 and EC2) (10). The glycosylation sites present in the EC2 loop contribute to protein variability (10, 11). The polar residues in the TM3 and TM4 domains play a role in intramolecular interactions (11). KAI1 is involved in cellular adhesion to the extracellular matrix by interacting with various integrin complexes that bind to fibronectin and laminin (12). Tetraspanins have the ability to signal through multiple protein families, including integrins, resulting in a diverse network with numerous functional and translational mechanisms (13). KAI1, which has such structural characteristics, has been studied as a variety of proteins capable of regulating cellular signal processes (6, 9, 13). In particular, based on recent reports, there are some differences in hematopoietic cells in BM of KAI1 knockout mouse compared with normal control (14, 15). Loss of KAI1 does not affect ST-HSC, MPP, or blood cell composition in mice, but significantly reduces the distribution of LT-HSCs (4). When hematopoietic cells from KAI1 KO mice are transplanted into other mice, it takes longer for the cells to migrate to the BM and there are increased numbers of differentiated blood cells such as CD45+ macrophages (16). This suggests that KAI1 plays a role in preventing LT-HSC differentiation in mice. And in KAI1 KO mice, the number and size of osteoclasts, that play a critical role in bone remodeling and maintenance of bone mass, was reduced. These indicate that KAI1 is important not only for the maintenance of LT-HSCs, but also for homeostasis of the BM-niche (15).

Previous studies have shown that KAI1 expression is restricted to the LT-HSC population within the bone marrow, but is not expressed in the more differentiated ST-HSC or MPP populations (4). Degradation of KAI1 protein has been shown to result in the loss of LT-HSC self-renewal ability and the onset of differentiation (4). Therefore, KAI1 has been shown to play a critical role in maintaining LT-HSC quiescence within the bone marrow niche, which is essential for the long-term maintenance of hematopoietic stem cells.

There are several different atypical chemokine receptor (ACKRs) that have been identified, each with its own unique expression pattern and ligand binding profile. Some ACKRs, such as ACKR2, are expressed on the surface of lymphatic cells and regulate the immune response in different pathological conditions, including infection, allergy, and cancer (17). Other ACKRs, such as ACKR3, are highly expressed on endothelial precursor cells (EPCs), which are a pivotal role in promoting angiogenesis, and regulate the survival effect of CXCL12 on EPCs and are related in CXCL12-mediated adhesion, proliferation, migration and tube formation of EPCs (18).

ACKR1, also known as Duffy antigen receptor for chemokines (DARC), binds over 20 different CC and CXC chemokines (19, 20). DARC is a seven-transmembrane domain protein mainly expressed on erythroblasts, vascular endothelium, and a subset of epithelial cells (21).

DARC is highly expressed in nucleated erythroblasts but also expressed on the lymphocytes (4), cancer cells (22, 23), and the surface of BM Mϕ (3, 4). Since DARC was not well known to be expressed in Mϕ, various techniques such as western blot or FACS, real-time PCR, and immunofluorescence analysis were employed on actual bone marrow tissue to confirm that DARC is indeed expressed in Mϕ (3). The exact number of DARC-positive macrophages in the bone marrow was analyzed using the image stream experimental technique, which allowed for direct visualization of fluorescence staining in individual cells (24). Similar to FACS analysis, each single cell population was designated by forward scatter and side scatter analysis based on scatter ratio (3). Subsequently, analysis was conducted by identifying cells positive for each marker, including DARC and BM Mϕ (CD11b+F4/80+) (3). About 2 to 3% of BM Mϕ expressed DARC on the membrane. This subtype of Mϕ is a rare population but is crucial for maintaining BM homeostasis.

Previous reports have shown that vascular endothelial DARC directly binds to KAI1 positive cancer cells to suppress tumor metastasis (25, 26). This interaction results in the inhibition of tumor cell proliferation and induction of senescence by regulating the expression of TBX2 and p21 (25). Interestingly, it has been found that direct binding between LT-HSC and macrophages is crucial to maintain KAI1 expression in the bone marrow environment (3, 4). Among the macrophage (Mϕ) population, a specific subtype of Mϕ present in the BM niche has been identified as a key cell that maintains the LT-HSC in a quiescent state, allowing it to maintain self-renewal by maintaining KAI1 expression (3).

DARC on the surface of BM Mϕ binding to KAI1 on the membrane of LT-HSC activates the TGF-β/Smad3 signaling pathway through PKC-α in LT-HSC. This signal induces CDK inhibitors such as p21, p27, p57 and suppresses cell cycle, and induces the dormancy of LT-HSC. Conversely, when the expression of DARC was decreased, the KAI1 level on the membrane of LT-HSC was decreased, eventually leading to cell cycle entry and differentiation (Fig. 1) (3, 4).

The gene editing approach is a powerful tool that allows researchers to make precise changes to the DNA sequence of cells or organisms. One approach to gene editing is to use site-specific nucleases, such as CRISPR-Cas9, to introduce targeted double-strand breaks in the DNA. When these breaks are repaired, it can result in the deletion or modification of the targeted gene. Thus this technique can target and destruct a specific gene (27-29). Using this technique, DARC gene was able to be specifically targeted and removed in monocytes and macrophages (3, 24), which enabled the more precise analysis of effects of DARC-positive macrophages on LT-HSCs. CSF1R (colony-stimulating factor 1 receptor) is a well-known gene that is only expressed in monocytes or macrophages (3). Tamoxifen-inducible macrophage-specific DARC knockout mice were generated by crossing mice having CSF1R promoter-driven Cre and mice having DARC-flox. By cutting out the DARC gene in the genome of mono/macrophages, it is possible to study the effects of DARC-positive mono/macrophages on LT-HSCs in vivo. When DARC-positive mono/macrophages were ablated by administration of tamoxifen, the percentage of LT-HSCs in G0 significantly decreased while the number of WBCs and neutrophils increased, suggesting that the awakened LT-HSC proliferated and differentiated (3).

Previous research has shown that mono/macrophages play roles in regulating the mobilization and quiescence of HSPCs in the bone marrow (30). Specifically, the α-smooth muscle actin (α-SMA)-positive BM Mϕ population, which is rare, induces cyclooxygenase-2 (COX2) and helps to prevent ROS production (30). Interestingly, DARC-positive macrophages belong to this rare α-SMA+ COX2+ population (3). Such a triple positive cell (DARC+ α-SMA+ COX2+) would be the perfect escort warrior to protect the LT-HSC from a harmful hematopoietic niche in BM. To maintain a low ROS environment for HSC homeostasis in the BM, COX2 of α-SMA+ macrophages generate prostaglandin E2 (PGE2), which prevents ROS production in LT-HSC and maintains the characteristics of stem cells (30). Notably, a newly discovered DARC-positive macrophage population (DARC+CD11b+F4/80+ α-SMA+COX2+) was found to regulate the cell cycle of LT-HSC. About 95% of DARC+ Mϕ express α-SMA and COX2 in the BM (3). In conclusion, among the macrophages in the BM, only a very rare 2% of the macrophage population bind to LT-HSC and regulate their quiescence, and among them, the triple-positive macrophages (α-SMA/COX2/DARC) prevent LT-HSC differentiation and protect the HSC population from the attack of ROS (Fig. 2).

Leukemia, one of the different types of cancer, affects the blood and bone marrow, and treatment options can vary depending on the type of leukemia and the patient’s individual circumstances. Some of the main treatment options that are used for leukemia are chemotherapy (31), targeted therapy (32), radiation therapy (33), immunotherapy (34) and stem cell transplantation (35). Stem cell transplantation, also known as a bone marrow transplantation, is to replace the patient’s diseased bone marrow with healthy stem cells from a donor. Stem cell transplantation may be an option for patients with more aggressive forms of leukemia or those who have relapsed after initial treatment (36).

Assuming an annual frequency of 84,000, 1.5 million hematopoietic stem cell transplantations (HSCTs) have been conducted since 1957 (37, 38). HSCTs are often employed in the clinic to treat a range of hematologic and non-hematologic malignancies, including leukemia, lymphoma, and multiple myeloma (4, 37). Among cell types of HSCTs, CD34-expressing HSCs are mainly used for the treatment of leukemia with limited specificity. CD34 is a transmembrane glycoprotein that is expressed on the surface of hematopoietic stem cells (HSCs) and progenitor cells, as well as on endothelial cells and some other cell types (39). CD34 is a marker that is used to identify and isolate these cells for transplantation and other therapies. CD34-positive cells are capable of differentiating into all of the different blood cell types, including red blood cells, white blood cells, and platelets (40). CD34 is also expressed on the surface of endothelial cells (41).

In hematopoietic stem cells, CD34 is expressed on the surface of short-term repopulating hematopoietic stem cells (ST-HSCs) and Multi-potent progenitors (MPPs) as well. Compared to ST-HSCs and MPPs, long-term repopulating hematopoietic stem cells (LT-HSCs), which have a relatively long-lasting ability for self-renewal and differentiation, do not express CD34. Instead of CD34, KAI1 is strongly expressed on the surface of LT-HSCs but not on the surface of ST-HSCs or MPPs (4). KAI1 interacts with DARC+, α-SMA+, COX2+ triple-positive macrophages to maintain the dormancy of LT-HSCs (3). When treating leukemia using HSCs, sorting out and amplifying the number of LT-HSCs that particularly express KAI1 is likely to provide better therapeutic outcomes than other approaches. There are limitations of current therapeutic approaches employed in the management of leukemia. Intensive treatment regimens such as chemotherapy to ablate BM cells including malignant ones is required (42). The following stem cell transplantation reconstitute BM with normal cells and is pivotal for the survival of patients recovering from pancytopenia and infection. To avoid severe and prolonged pancytopenia after BM ablation, they may choose milder chemotherapy, which may lead to relapse following remission (43). Considering that rapid and complete BM reconstitution is very important, transplantation of HT-HSCs based on KAI1 selection would be more effective than the current CD34-based cell selection. In this context, it becomes crucial to develop a technique that enables the substantial expansion of LT-HSCs while maintaining their potential. The isolation of LT-HSCs poses a formidable challenge (44), because when we try to expand the number of LT-HSCs we have to lose KAI1 expression and stemness (45). Cyclic ex vivo culture of LT-HSCs between dormancy with maintenance of KAI1 and proliferation with loss of KAI1 may be the solution to expand number of LT-HSCs ex vivo before transplantation. Or the strategy to combine LT-HSCs expressing KAI1 and macrophages expressing α-SMA+COX2+DARC+ would be another option to improve the therapeutic efficacy of stem cell transplantation for hematological disorders (Fig. 3).

The authors have no conflicting interests.